HN

Comments (10)

• jacomyma
  Fantastic article.For readers interested in this, let me point to the somewhat similar "Activation Atlas" interactive paper, published in the sadly now-defunct scientific journal Distill.https://distill.pub/2019/activation-atlas/
• markusMB
  Beautiful illustrations I find, 'Playing' is just the free and motivated version of 'exploration'.One thought on your nicely illustrated "key observation [is] that neural networks tend to place features along directions": my guess is that the neural net was TOLD to behave that way by choosing e.g. Cosine Loss?
• Menu

  archermarks
  Nice article! The generated images make me so nostalgic for the early days of AI image generation. DeepDream and others had such uncanny, interesting generations.
• RealityVoid
  For some reason, the uncanniness of the feature pictures are deeply unsettling for me. It just stirs intense unease. A bit amusing, to be honest.
• joaquincabezas
  This article is very well structured and provides just the right amount of details for non-practitioners to enjoy it.Mechanistic interpretability is a fun topic to "play with" (good title there). I recommend watching videos featuring Neel Nanda or Chris Olah
• Menu

  jcattle
  Very nice visualizations, thanks for that!One thing I still struggle with in my head is how these vision embeddings can then be used to give LLMs eyes.Because you somehow need a giant training set which describes images in natural language, no? Is that actually how it works, or is there some smart trick so you don't need to pay labellers a bunch of money to look at pictures and describe them.
• agentbraker
  Awesome project! Preserving and sharing knowledge like this is incredibly valuable. Thanks for making these resources accessible to everyone.
• anon
  undefined
• anon
  undefined
• cdogukank
  [flagged]
• SkitterKherpi
  [dead]